Exactly what does "failed writing" mean?

[jtara92101]

I've seen this occasionally with various devices. When I do see it, it happens consistently.

 

Does this mean the EPROM on the device has gone bad? Or does it mean something else?

 

(Aside: is there some comprehensive guide to what these error message in activity log actually mean?)

 

The ISY is definitely communicating with the device, because I consistently see 1 hop left.

 

I'm going to replace this device in any case, as the airgap tab is broken, which makes it very inconvenient to set-up, and I do want the airgap to work in case the thing goes completely on the fritz - it's good to be able to disable a dimmer quickly without having to shut off the circuit!

[stusviews]

That usually indicates communication errors. A loose connection can cause that. Sometimes the error is an unrelated glitch on the power line. One hop left is OK, but not great.

[jtara92101]

Confused about hops now.

 

I've never seen more than "2 hops left". Even with devices plugged into the same duplex outlet, or wire-nutted together.

 

So, I assume if no repeating, then that's considered 1 hop, and there are two left.

 

If the device is on the other phase, it will have to use a hop, then you have 1 hop left.

 

So, "it's not great" if the device isn't on the same phase. But it's impossible to do better, save from re-wiring your house all on the same phase, or using dual-band devices, and all within range of every other device.

 

Am I wrong about this?

[jtara92101]

Absolutely bizarre! (As ALWAYS...)

 

I linked a KPL button to a LampLinc. "Failed writing. YET IT WORKS.

 

But at the same time, the KPL gives an error indication.

 

I press the C button. It turns on the LampLinc. Then it blinks the Off button indicating an error.

 

I press the C button again. It turns off the LampLinc. And again blinks the Off button indicating an error.

 

I can also turn the LampLinc on and off from the ISY.

 

I can see on the ISY console, ISY thinks there's some writing yet to do to both the KPS and the Lamplinc.

 

Yet it works just fine (save for the annoying blinking). And the button did not turn the LampLinc on/off before I linked it. (So I did not have old links left.)

[stusviews]

I press the C button. It turns on the LampLinc. Then it blinks the Off button indicating an error.

 

A KPL secondary button does not have an Off button and an On button, so I don't know what, "it blinks the Off button" means.

 

Do you have a ! or 1011 next to the KPL?

[larryllix]

Devices always do three hops. Hops = repeats, and the original signal send and receive, doesn't count.

 

What you see is three hops less the number before the ISY sees the signal from the device or vice versa.

 

1 hop left is not a bad thing if there is a lot of distance bewteen the device and the PLM. This is a natural occurence with Insteon devices and the way they are supposed to work. Varying Hops Left can be a bad sign, though.

 

You have to judge if the signal should have hopped through as many devices as reported. If the device is close to the PLM you should have three hops left.

 

A writing device error could occur due to signal not getting there, device defect or unplugged, or ACK not coming back from the device. Of course Insteon usually repeats the signal try 3 times on top of all that.

[jtara92101]

The large OFF button of the KPL blinks. Actually, I think it is the large ON and OFF buttons alternately blink.

 

I have observed this MANY times throughout the years.

 

There was no ! next to the devices. Only the 1010 (or whatever) indicator showing there was something the ISY needed to write.

 

I re-did "write changes to device" and this time it "took".

 

As always a mystery.

 

As well, no more blinky.  Just works as it should.

[larryllix]

The large OFF button of the KPL blinks. Actually, I think it is the large ON and OFF buttons alternately blink.

 

I have observed this MANY times throughout the years.

 

There was no ! next to the devices. Only the 1010 (or whatever) indicator showing there was something the ISY needed to write.

 

I re-did "write changes to device" and this time it "took".

 

As always a mystery.

 

As well, no more blinky.  Just works as it should.

Comm problems usually.

Watch the Hops left for many operations and see if it varies.

[jtara92101]

And now the KPL is dead as a doornail. Doesn't light up, switches don't control linked devices, does absolutely nothing at all.

 

About to just go to Home Depot and get a bunch of standard dimmers.

 

Same as the one I replaced.

 

Seems sometimes gets powered. Sometimes doesn't. I don't think it's a wiring issue (unless inside the device itself.). The switchbox powers everything else in the room. Which works.

[jtara92101]

I'll bet it's the &^%$! air-gap. Second one in a row. They just go bad after a while. And, or, there is an issue with insufficient mechanical clearance, once you put a faceplate on it. Button it up, and stops working.

[jtara92101]

Failed air-gap. I can see it. It doesn't close all the way.

[stusviews]

A loose connection the the Insteon device can cause that. The air-gap is independent of the functioning of the switch. Yes. you can use the air-gap to cut power to the switch, but that's not the part that goes bad, only the tab breaking off. The device can function with a broken air-gap.

 

When you "button up" a device and it stops working, then that's a clear indication of a poor connection.

[larryllix]

Have you got Hot, Neutral and Load wired to this KPL?

This sounds a lot like you have the fixture in series with the KPL, like you could with a mechanical switch.

[jtara92101]

Photos later, once they sync. Definitely a small gap with it supposedly closed.

 

I pull the air-gap. Big gap. I push it back in. Smaller gap. Mechanical failure. Some springy failed, deformed, whatever.

[larryllix]

How are you seeing an air-gap. Mine are opaque and there is nothing I can see.

[jtara92101]

Mine are clear. I thought they were all clear?

[larryllix]

Mine are clear. I thought they were all clear?

Been a long time. I may be wrong on that one.

 

Is your KPL black wire connected to permanently live hot?

[stusviews]

There are no springs in the air-gap, only a detent. If you can pull the air-gap out and it stays in place, then the detent is working. If the device has power when you remove it from the box (still connected), then the air-gap is working.

[jtara92101]

See attached. The air-gap is is NORMAL position. You can see the gap. The gap is much wider when you pull it out.

 

It doesn't have springs, per-se. But presume those silver-colored parts that the contacts are attached to are spring steel. You need SOMETHING springy to keep any contacts together, unless they are very heavy, rely on gravity, and only installed in specific orientation.

 

All a bit hard to see in the resized photo. It's quite clear in the original. You can see that the plastic detent pulls up (the switch is upside-down in the photo) on the upper (lower) spring-steel part, pulling the contacts apart.

 

I suppose I might fix it by sticking something in one of the holes directly above each spring-steel piece. I don't think the holes are for that purpose, as the case is just symmetrical. The wires exit from similar holes at the top of the case. (The switch is upside-down in the photo.

[mwester]

Yep, that one's broken.

 

I had a switchlinc that did that.  I used a needlenose pliers and some care, and bent the contacts so that they worked properly, then put it back into service.

[jtara92101]

Well, that was easy!

 

I fixed the Keypadlinc. Didn't even have to take it out of the box.

 

Turned off the power, and pushed through one of the little holes with the tip of a meter probe to push on the spring-that-doesn't-exist

 

That deformed it enough so now it makes contact again.

 

I think that the airgap should be considered a one-use (or n-use, where n is a small number) device. LIke old Watts multi-turn (globe valve) angle stops. (My NEXT task. At least the KPL isn't a disaster about to happen. 

 

Makes me wonder about the placement of those holes, and how they were justified to UL - because they are calling out to consumers to stick things in them to fix flawed airgaps.

 

- are they there for assembly flexibility? (Housing might be used for different devices, some might have wires coming out the bottom)

 

- are they there for ventilation?

 

- are they there as access to test points? (You could easily probe both sides of the air gap. And if you "accidentally" push to hard, you might accidentally fix a broken device

 

Now I wonder if there actually was any real communication issue. Was the airgap just "wigging out" momentarily causing it to lose power? (It wasn't within eyesight, in another room from my screen with ISY console.) I do think that's unlikely, but it was firmly mounted at the time, and so there's an oft-chance vibrations in the wall/building...

 

I had airgapped it to reset at start of install. In the future, I will not airgap. Half the time now I don't anyway. My collection of devices has proven so flakey that I am remiss to go to the trouble of putting them in the wall until they are proven to work and any linkages made. So, I put them on my zip-cord setup, and it is easy to reset by holding-in the reset button whilst plugging-in the cord to an outlet. It is both easier than grabbing the little handle, and now I know it also will save wear and tear on a limited-cycle component.

[jtara92101]

Well, the device is still flakey communicating, and will have to be replaced in any case. Maybe it has multiple failures. Maybe it was in a bathroom before I moved...

[jtara92101]

Doh!

 

Do Insteon devices flash an error indication or have some communication failure mode when the is no load plugged-in to the (or "a", if serveral) responder(s)?

 

I plugged my bedroom lamps back into the LampLinc that is responder for the main on/off of the KPL in question. Now all seems to work without blinky lights.

 

Maybe the load acted as an interference-sucker. That would be the opposite of a signal-sucker.

[Teken]

This clear case with the multi holes has been a default housing for years. The wires come out depending upon assembly person and the model year PCB. As older hardware had different component layout necessitating the wires to exit at random points.

 

In the big picture the wires could have come out the same hole each time simply by using longer wire.

 

But it appears to me anyways the wires are precut and thus its pushed in where ever the wires will be the same length.

 

Also most if not all of the air gap have a return spring. It should be noted there was one model which the air gap wasn't present and pulling on it would damage the unit.

 

 

=========================

 

The highest calling in life is to serve ones country faithfully - Teach others what can be. Do what is right and not what is popular.

[jtara92101]

I just pulled a SwitchLinc from my box-o-dimmers, a "newer" one (for me) v5.15. (Not dual-band though).

 

This one has a quite a different airgap design. There is indeed a real spring, rather than relying on spring steel holding the contact points.

 

And the holes above the airgap are missing as well. But my repair would not be useful with this design, anyway.

 

Seems a much more reliable design as there is a spring pulling on the plastic gizmo and putting tension on the upper arm so as to pull the contact points together much more positively.